Lower Airway Dysbiosis in Nontuberculous Mycobacterial Lung Disease Drives a Neutrophil Extracellular Trap-endotype and Lung Injury

Le résumé fourni par la source

Abstract RATIONALE: Pulmonary dysbiosis is associated with higher neutrophil extracellular trap (NET) levels in Bronchiectasis. The WILLOW trial reported a reduction in exacerbations with Brensocatib (cathepsin-C inhibitor that can disintegrate NET), thus identifying NET as a treatable trait, but the study did not include NTM+ patients. Ex-vivo studies have shown that NTM infection can also drive NET formation. Here, using ∼200 patient-derived bronchoalveolar lavage fluid (BALF) samples and a preclinical mouse model, we investigate if NTM infection in dysbiotic airways can drive NET-mediated lung inflammation. METHODS: BALF samples were collected from bronchiectasis subjects (108 NTM-, 92 NTM+) undergoing clinically indicated bronchoscopies. Samples underwent 16S rRNA sequencing (analyzed using QIIME2, PhyloSeq, EdgeR); and a NET ELISA assay. Lower airway metacommunities were identified with Dirichlet Multinomial Mixtures (DMM); microbe-microbe associations were analyzed with co-occurrence analyses and microbiome-NET association was analyzed with Microbiome Multivariate Association with Linear Models (MaAsLin). For murine studies, lower airway dysbiosis was induced by intra-tracheal instillation of “mixed oral commensals” (MOC) (mixture of anaerobic bacteria), followed by aerosol NTM infection. Readouts included BALF NET ELISA, flow cytometry, colony forming units (CFU) and histopathology. RESULTS: Dirichlet Multinomial Modeling (clustering analysis) identified three microbial metacommunities that were unique in bacterial load, microbiome diversity and composition, with significant differences in NET levels (Fig. 1A, p<0.001). NET levels were higher in NTM+ subjects. Mycobacterium had a synchronous occurrence with oral commensals such as Chryseobacterium, Capnocytophaga and Fluviicola; and neutrophils (Fig. 1B). MaAsLin2 analysis showed that Mycobacterium also associated with NETs (coefficient of association 0.18) (Fig. 1C). In the murine model, NTM and MOC+NTM drove NET formation, with significantly higher NET levels than control (Fig. 1D). Flow cytometry was performed over 2-6 weeks and demonstrated an increase in neutrophils (data not shown) and other cells that can drive NET formation as well as a pro-inflammatory profile such as Th17+, γδT+ and PD-1+cells in the MOC+NTM group (Fig. 1E-G). Mouse lung CFUs and histopathology were consistent with lung inflammation. CONCLUSIONS: In this study, we identified that lower airway metacommunities had different NET levels, and in NTM+ subjects, Mycobacterium had a synchronous occurrence with oral commensals as well as with NETs. Using a bedside-back-to-the-bench paradigm, we confirmed in our murine model that NTM drives NET formation and lung inflammation/injury in dysbiosis. Our study suggests a pathogenic role for NET in NTM-LD and future studies could help identify a NET-predominant endotype as a treatable trait in NTM-LD.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.